There's always more than one way to do it .. I like to switch things around and test for the least likely outcome first .. check to see if the year is divisible by 400, then 100, and finally 4. I think it also makes the code a little more transparent, but everyone's got their own style ..

#!/usr/bin/perl

use strict;
use warnings;

my %testData = (
    1900 => 0,
    1904 => 1,
    1972 => 1,
    1973 => 0,
    1999 => 0,
    2000 => 1,
    2001 => 0,
    2004 => 1
);

{
    foreach my $this ( keys %testData ) {

        if ( $testData{$this} == leapYear($this) ) {

            print "Correct - $this year "
              . ( $testData{$this} ? "is" : "is not" )
              . " a leap year.\n";
        }
        else {

            print "Error with $this entry ..\n";
        }
    }
}

sub leapYear {

    my $year = shift;

    if ( $year % 400 ) {
        if ( $year % 100 ) {

            #  3. Simple case: it's a leap year if it's divisible by 4
            return ( $year % 4 ) ? 0 : 1;
        }
        else {

            #  2. If it's divisible by 100: it's not a leap year.
            return 0;    
        }
    }
    else {

        #  1. If it's divisible by 400: it's a leap year.
        return 1;        
    }
}
[download]

I've numbered the comments to make it a little clearer which result might be returned first, although this could be made even clearer by reversing the meaning of the conditionals so that the routine looks like this:

sub leapYear {

    my $year = shift;

    if ( $year % 400 == 0 ) {

        #  1. If it's divisible by 400: it's a leap year.
        return 1;        
    }
    else {
        if ( $year % 100 == 0 ) {

            #  2. If it's divisible by 100: it's not a leap year.
            return 0;    
        }
        else {

            #  3. Simple case: it's a leap year if it's divisible by 4
            return ( $year % 4 ) ? 0 : 1;
        }
    }
}
[download]